Identifying risk factors

Experts have identified certain factors that heighten the risk of violence among people with and without psychiatric disorders. As a general matter, mental illness correlates with a risk of violent acts far less than do male gender, youth, and social or economic disadvantage. Other clear risk factors for externally directed violence include a history of serious abuse or other trauma as a child and violent victimization over the life course. In addition, episodes of recent violence, prior arrests, certain personality disorders such as psychopathy, conduct disorders, high levels of anger, and the presence of escalating violent fantasies all increase the likelihood of violence. According to a recent article by experts in behavioral health, public health and gun violence, ―[e]vidence from studies in criminology and developmental epidemiology has shown that risk factors for crime and violence are similar in persons with mental illness and the general population, and that risk exposure often begins early in life.‖ (Swanson, JW et al., ―Mental illness and reduction of gun

violence and suicide: bringing epidemiologic research to policy,‖ Annals of

Epidemiology (2014).) Although some variations emerge with a closer look at particular types of violence, these factors remain salient for violence across the board and homicide in particular. For example, men perpetrate more than 90% of homicides in the United States. Men also complete the act of suicide approximately four times as often as women do, many of those by firearm. The vast majority of school shooters and perpetrators of other mass killings have been male.

We also know that substance abuse has a stronger association with acts of violence than does a psychiatric diagnosis. Substance abuse also combines with mental illness to

increase the risk of violence

significantly. In what has been called the best epidemiological study ever conducted on rates of violence among

“What we know about violence is that it’s dynamic. So every single one of us is capable of engaging in violence under the right circumstances. If our lives are threatened, a family member’s life is threatened, we are capable of acting violently.”

Marissa Randazzo, Ph.D., Sigma Threat Management Associates, Testimony presented to the Sandy Hook Advisory Commission, March 22, 2013.

158

populations with and without diagnosed mental illnesses, persons who abused alcohol were more than twice as likely to commit acts of violence than those diagnosable with a major mental disorder who did not abuse alcohol, and those who abused drugs were nearly three times as likely to engage in violence. (Swanson JW et al., 1990. Violence and psychiatric disorder in the community: evidence from the Epidemiologica Catchment Area surveys, Hosp. Community Psychiatry 41(7):761-770.) Another major study showed that where persons living with mental illness were not abusing drugs or alcohol, they had no statistically significant differences in their rates of violence compared to the general public. Alcohol and drug abuse are highly salient risk factors for violence, and when a person with a psychiatric condition has a simultaneous substance abuse problem, the risk of violence escalates. (MacArthur Violence

Risk Assessment Study (MVRAS), summary available at:

http://www.macarthur.virginia.edu/violence.html.) Substance abuse

frequently co-occurs with symptoms of mental disorder, particularly when those symptoms have not received adequate treatment. Drugs and alcohol likely account for much of the increased risk of violence among individuals with psychiatric illnesses. Therefore, efforts to address the contribution of mental illness to violence generally must confront substance abuse to have any measurable effect.

We also know that rates of gun violence in general, and particularly gun fatalities, correlate strongly with higher rates of gun ownership. Whereas the United States has both extremely high rates of gun ownership and high rates of firearm-related deaths, Japan and the United Kingdom have very low gun ownership rates and correspondingly low rates of gun-related deaths. In a recent study evaluating the relationship between rates of gun ownership and deaths by firearms across twenty-seven developed countries, researchers found ―a significant positive correlation between guns per capita per country and the rate of firearm related deaths, with Japan being on one end of the spectrum and the US being on the other.‖ Their statistical analysis identified a far weaker correlation between rates of mental illness – estimated by looking at

major depressive disorder – and gun deaths, and no significant correlation at all between rates of mental illness and general crime. (Bangalore & Messerli,

―Gun Ownership and Firearm-related Deaths,‖ The American J. of Medicine,

Oct. 2013.)

Some countries have dramatically reduced their rates of gun violence with heightened regulation of firearms. For example, Australia implemented strict gun control legislation across its states and territories following the 1996 murder of thirty-five people at a popular tourist destination by a young man with assault weapons. Since then, Australia has seen a significant reduction in firearms-related deaths. The new law banned all automatic and semi- automatic long guns (establishing national buybacks of semi-automatic rifles, self-loading and pump-action shotguns, and handguns) and instituted a strict licensing and registration requirement for all legal firearms. There have been no mass shootings in Australia since 1996, and the firearms mortality rate has decreased from approximately .27 per 100,000 to .13 per 100,000 in recent years. Viewed from a different angle, death by firearm is now well over twenty times (and as much as twenty-seven times) more likely to occur in the United States than in Australia. As the Child Advocate‘s report observed, ―[t]he conclusion that access to guns drives shooting episodes far more than the presence of mental illness is inescapable.‖ (OCA Report at 79.) In addition to declining rates of gun homicide and elimination of mass shootings altogether, rates of suicide by firearm also saw striking declines in Australia during this period. (Peters, R. (2013), ―Rational Firearm Regulation: Evidence-based Gun

Laws in Australia,‖ in Daniel Webster & Jon Vernick, eds., Reducing Gun

Violence in America: Informing Policy with Evidence and Analysis (Baltimore: Johns Hopkins Press),195-204; Alpers, P. (2013), ―The Big Melt: How One Democracy Changed after Scrapping a Third of Its Firearms‖ in Webster & Vernick, 205-211).

One exception to the weak correlation between psychiatric disorder and violence involves individuals undergoing a first psychotic episode, particularly if high levels of anger accompany persecutory or paranoid delusions. In

160

testimony presented to the Commission, Dr. Madelon Baranoski of the Yale School of Medicine offered a useful account of why particular symptoms of psychosis may increase the risk of violence. A person suffering from a psychotic illness often confuses internal thought processes with real events in the world. So if that person is feeling scared, he or she may mistake thoughts

of danger for actual danger. Likewise, internal feelings of anger and unease

may appear to be coming from external sources. Persecutory delusions, such as beliefs that a person is being spied on or is the target of a conspiracy, can be extremely frightening for someone who cannot reliably distinguish his or her own thoughts from external reality. Yet even such delusions of persecution do not generally appear to lead to violence in the absence of unregulated anger. High levels of anger increase the risk of violence across our society, whether or not psychiatric illness is present. (Baranoski, April 12, 2013 testimony.)

Some researchers have estimated that a relatively significant risk of violence – several times the risk for the general public – accompanies a first episode of untreated psychosis, especially during the period between the onset of illness and the beginning of treatment known as the duration of untreated psychosis, or DUP. A large-scale analysis of the available research on homicides committed by persons with a psychotic illness found that annual homicide rates by individuals with untreated psychosis were approximately fifteen times higher than rates for individuals with treated psychotic illnesses. (Nielssen, O. & Large, M. (2010), ―Rates of homicide during a first episode of psychosis and after treatment: A systematic review and meta-analysis,‖ Schizophrenia Bulletin, 36: 702-712). For those who had received mental health treatment, total homicide rates were approximately 1 in 10,000 annually. For those in a first episode of psychosis before treatment began, however, homicide rates approached 1 in 700.

A very recent study out of Sweden suggests that a diagnosis of schizophrenia or related psychotic disorder is associated with an increased risk for violent offenses, suicide and premature mortality. Researchers also found,

though, that specific risk factors beyond the diagnosis enhance the likelihood

of these poor outcomes. Published in The Lancet, this large-scale study

compared records of over 24,000 Swedes diagnosed with schizophrenia to those of their siblings without such diagnoses and of the general population. It isolated three risk factors that typically increase the rates of all three adverse outcomes: drug use disorders, a history of violent criminal behavior, and self- harm. (Seena Fazel et. al. (June 2014) ―Violent crime, suicide, and premature mortality in patients with schizophrenia and related disorders: a 38-year total

population study in Sweden,‖ The Lancet Psychiatry, 1:1 at 44-54.) Once

again, this study supports the conclusion that a specific diagnosis alone tells us very little about a person‘s likelihood of engaging in acts of violence. Instead, we must attend carefully to the particular risk factors that research conducted throughout the world has linked to violence and self-harm.